AI System (legal definition)
Under the EU AI Act: a machine-based system that operates with some autonomy and infers outputs from inputs.
Article 3(1) defines an AI system as machine-based, designed to operate with varying levels of autonomy, possibly adaptive after deployment, which infers from its inputs how to generate outputs such as predictions, content, recommendations, or decisions that can influence environments. The definition is deliberately technology-neutral, and ‘infers’ is the operative word — it is what separates an AI system from ordinary software. The Commission issued guidelines in February 2025 to help draw that line.
In practice: A rules engine written by hand is likely out of scope; a trained scoring model is in.
Plain-English summary of Article 3(1). The binding text is Regulation (EU) 2024/1689.
Where this comes up
- AI Consultant Certification in 2026: Skills, Training, and Career Path
- AI for Financial Advisors in 2026: Client Service, Research, and Compliance-Aware Workflows
- AI for Personalized Learning: Enhancing Educational Experiences
- AI for Retail in 2026: Use Cases, Workflows, Tools, and Training
- An AI Just Designed an AI Chip in Two Weeks: What Redwood Really Proves
- Anthropic's Model Hardware Standard: AI Agents Step Into the Physical World